JP2005223503A - Video signal processing apparatus and video signal processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the technology of carrying out excellent sequential scanning conversion by preventing image quality deterioration even when regularity of a sequence is disturbed because of the occurrence of a deviation in 2-3 pull-down phase due to an edit point or the like in the midway of a telecine video signal. <P>SOLUTION: The video signal processing apparatus detects a correlation between a target field and the field of one preceding frame of the target field on the basis of difference information between frames, and a telecine sequence detection means 21 discriminates whether or not a video signal is the telecine video signal depending on the result of detection. When a field different from a field at a particular phase caused once per 5 fields is a still picture field, a sequence deviation detection means 22 detects it, the result of detection is utilized for the discrimination in the telecine sequence detection means, and when deviation in the 2-3 pull-down phase takes place, the operation of carrying out IP conversion on the basis of the detection of the telecine video signal is temporarily released. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention is an image signal processing apparatus for generating an interpolated scanning line from an interlaced video signal and converting it into a progressive video signal. Video signal processing apparatus and video signal processing method for detecting whether or not the interlaced scanning signal converted in step 1 is a video signal subjected to telecine conversion in the process of converting the interlaced scanning signal into a video signal for sequential scanning. It is about.

Movie films are shot at 24 frames (frames) per second, whereas standard television signals such as NTSC signals and high-definition signals are 60-field video signals per second for interlaced scanning. ing. Therefore, when a signal such as a movie film is broadcast as a television broadcast program or packaged as a video package, a signal of 24 frames per second is converted into an interlace signal for television broadcast of 60 fields per second. become. In this telecine conversion, generally, one frame is converted into 2 fields and the next frame is converted into 2-3 pull-down so as to correspond to 3 fields, and once every 5 fields, the same field as 2 fields before is repeated. By doing so, it is converted into an interlaced signal (hereinafter referred to as a telecine video signal) of 60 fields per second.
Note that the case where the odd-numbered frame is converted to 2 fields and the even-numbered frame is converted to correspond to 3 fields is called “2-3 pull-down conversion” in a narrow sense, and the odd-numbered frames are converted to 3 fields. The case of converting even-numbered frames so as to correspond to 2 fields is sometimes referred to as “3-2 pull-down conversion”, but these are substantially the same. This is called “conversion”.

  On the other hand, in recent years, when an interlaced television signal is displayed with high image quality, a process of converting the interlaced signal into a progressive scan signal (hereinafter referred to as IP conversion) is performed. In this IP conversion, in general, when a video is a still image (motion is equal to or less than a predetermined amount), interpolation (inter-field) is performed by fitting pixels of corresponding scanning lines in temporally adjacent fields. It is known to perform motion adaptive IP conversion in which an interpolation signal is generated by pixels in adjacent scanning lines in the same field (intra-field interpolation) in the case of a moving image (motion is greater than a predetermined amount). ing. Further, when performing IP conversion of an interlace signal that is a telecine video signal as described above, whether or not the video signal to be processed is a telecine video signal and a phase of 2-3 pulldown (formed by 2-3 pulldown). (Information indicating the number of the first field in the sequence consisting of five fields), and for the telecine video signal, a field converted from the same frame is used according to the phase of 2-3 pulldown. By restoring the progressive scan signal before the 2-3 pull-down conversion, it is converted into a progressive scan video signal.

  In the conventional video signal processing apparatus that detects and converts the telecine video signal to the IP, in the case of the 2-3 pull-down telecine video signal, the same field as the previous two fields is repeatedly converted once every five fields. Uses the characteristic that there is a still image field in which the difference information is zero or less than a predetermined value in one cycle in five fields according to the difference with the field one frame before (that is, two fields before) Thus, a video signal processing apparatus that detects a telecine video signal and performs IP conversion of the telecine video signal is shown. (For example, refer to Patent Documents 1, 2, and 3.)

  In addition, in order to prevent erroneous detection in telecine detection and to perform stable IP conversion of a telecine video signal, the number of screen matches in a telecine video signal sequence repeated in five fields is counted, and whether or not the telecine video signal is determined by the count value Some of them determine whether or not (see, for example, Patent Document 4).

  Further, a repetitive sequence for every five fields of interframe differences in a telecine video signal is detected, and a predetermined pattern deviation that is not a telecine video signal is detected from difference information in five fields (for example, Patent Document 5). (Refer to the previous page) Sequentially determine the pattern up to 5 fields before in each field, obtain a signal indicating the scene change point from the outside or input signal, and deviate from the predetermined pattern considering the signal indicating the scene change point In some cases, the telecine video signal is accurately detected (see, for example, Patent Document 6).

Japanese Patent No. 3398396 (FIGS. 1 and 2) JP 2002-330408 A (FIGS. 1 and 2) JP 2002-335441 A (FIG. 1) JP 2002-204433 A (FIG. 1) JP 2003-78926 A (FIGS. 1 and 2) Japanese Patent Laid-Open No. 2003-17984 (FIG. 1)

  However, in the conventional method, when the regularity of the 2-3 pulldown field sequence is disturbed by editing for broadcast, insertion of CM or telop, etc. in the middle of the telecine video signal, the 2-3 pulldown field sequence is used. Therefore, until the normal 2-3 pull-down phase is restored, IP conversion is performed by the interpolation signal generated using the signal in the wrong field, and a shift occurs in the image. As a result, image quality deterioration such as the occurrence of a problem occurs, and the IP conversion of a good telecine video signal cannot be performed.

  In the present invention, when the 2-3 pull-down field sequence is disturbed, it is detected reliably, and the result of interpolation by the motion adaptive interpolation processing means is used instead of the interpolation using the telecine conversion video signal. Therefore, it is an object to suppress image quality deterioration.

The present invention
A video signal processing device that generates an interpolated scanning line from an input interlaced video signal and converts it to a progressive video signal,
Field delay means for delaying a video signal by a plurality of fields in a field unit;
A correlation between the video signal of the field of interest and the video signal of the field one frame before the field of interest is detected, and the field of interest is a still image field or a moving image field based on the strength of the correlation between frames. Correlation determination means for determining whether or not
Based on the output signal from the correlation determining means, a field sequence of telecine conversion in a video signal telecine-converted by the 2-3 pull-down method is detected, and the input video is detected from the detection result of the field sequence of the telecine conversion. Telecine sequence detection means for determining whether the signal is a telecine converted video signal;
Based on the output signal from the correlation determining means and the detection result of the telecine sequence detecting means, the field is different from the field of the specific phase to be the still image field in the telecine conversion field sequence detected by the telecine sequence detecting means. Sequence deviation detecting means for detecting this when the is a still image field,
Scanning line interpolation means for switching processing for generating an interpolation scanning line signal based on the detection result from the telecine sequence detection means,
The telecine sequence detection means provides a video signal processing apparatus characterized by determining whether or not the video signal has been telecine converted using the output of the sequence deviation detection means. .

  According to the present invention, when a field different from the field that is the same as the previous two fields once in five fields in the telecine video signal is a still picture field, this is detected and the phase of the 2-3 pulldown field sequence is detected. By detecting the shift of the sequence, when the phase shift of the sequence is detected, the operation of performing the IP conversion based on the detection of the telecine video signal is canceled. Even when the regularity of the field sequence is disturbed, it is possible to stably detect the telecine video signal, prevent deterioration of the image quality in the IP conversion of the telecine video signal, and perform a good IP conversion of the telecine video signal. effective.

  The video signal processing apparatus and the video signal processing method according to the embodiment of the present invention detect a telecine video signal from difference information from a field one frame before and the same as two fields before once every five fields. When the field different from the field is a still image field, this is detected, and based on the detection result, the operation of performing IP conversion based on the detection of the telecine video signal on the basis of the 2-3 pull-down phase shift is temporarily performed. It is characterized by being released.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a configuration of a video signal processing apparatus according to Embodiment 1 of the present invention. The video signal processing apparatus of FIG. 1 is also used for implementing the video signal processing method of Embodiment 1 of the present invention.

  The video signal processing apparatus shown in FIG. 1 receives the interlaced scanning video signal Ia supplied to the input terminal 100, outputs it to the first terminal 101 as the actual scanning line signal R, and outputs the interpolated scanning line signal I. This is generated and output to the second terminal 102, and these are combined by the synthesis circuit 9 to generate a video signal PR for sequential scanning and output from the output terminal 103. Such interlaced scanning is performed. In the conversion (IP conversion) from the video signal to the progressive scanning video signal, it is determined whether the input interlaced scanning video signal Ia is a telecine video signal satisfying a predetermined condition, and the determination result Accordingly, IP conversion is performed by performing any of different interpolation processes.

  As shown in the figure, this video signal processing apparatus includes a timing signal generating means 1, first and second one-field delay means 2 and 3, and a motion detecting means 4 for detecting motion based on the interframe difference of the video signal. In accordance with the output of the motion detection means 4, a motion adaptive interpolation calculation means 5 that performs scanning line interpolation processing according to the motion, a telecine detection means 6 that detects whether the input video signal is a telecine video signal, and a telecine detection means 6 Telecine interpolation processing means 7 for performing scanning line interpolation processing corresponding to the 2-3 pulldown signal based on the signal from, and motion adaptive interpolation calculation means 5 according to the detection result of the telecine video signal from the telecine detection means 6; The selection means 8 for selecting and outputting one of the signals from the telecine interpolation processing means 7, the output of the first one-field delay means, and the output of the selection means 8 are double-speed converted. Combination alternately switched for each scanning line, and a synthesizing means 9 for outputting a video signal PR progressive scan.

  The timing signal generating means 1 separates the vertical and horizontal synchronizing signals from the interlaced video signal input from the input terminal 100 and generates a timing signal based on the separated synchronizing signals. For example, based on the vertical synchronization signal, a field pulse Vp indicating the timing for each field and an area signal te for indicating a predetermined area in one field are generated.

  The 1-field delay means 2 delays the interlaced video signal input from the input terminal 100 by one field, and the 1-field delay means 3 further delays the signal from the 1-field delay means 2 by one field. These combinations delay the video signal by a plurality of fields in a field unit.

  Using the input signal Ia from the input terminal 100 as the current field signal, the output signals Ib and Ic of the 1-field delay means 2 and 3 may be called 1-field delay signal and 2-field delay signal, respectively, and will be described later. For this reason, the signal supplied to the input terminal 100 is called the video signal (Ia) of the field of interest, and the output of the first one-field delay means 2 is called the video signal (Ib) one field before the field of interest. The output of the second one-field delay means 3 may be called a video signal (Ic) two fields before the field of interest, that is, one frame before.

  The motion detection means 4 receives the current field signal Ia, which is an interlaced video signal from the input terminal 100, and the two-field delay signal Ic from the one-field delay means 3, and obtains difference information between the signals in the two fields. Then, based on the obtained difference information, the motion of the video signal between one frame is detected, a motion detection signal indicating the motion of the signal is output, and sent to the motion adaptive interpolation calculation means 5.

  The motion adaptive interpolation calculation means 5 is based on the motion detection signal output from the motion detection means 4 for the 1 field delay signal from the 1 field delay means 2 and the 2 field delay signal from the 1 field delay means 3. Performs adaptive IP conversion. Interpolation scanning line signal generation processing is performed using the 1-field delay signal (Ib) as an interpolation target. Therefore, based on the motion detection signal, temporally adjacent fields, for example, the temporally previous 2-field delay signal Inter-field interpolation by inserting pixels of the corresponding scanning line from (Ic), or in a field from a video signal in the vicinity of the interpolation target field (one-field delay signal Ib), for example, a pixel signal positioned above and below in the vertical direction Interpolation processing is performed, and an interpolation signal Im in the interpolation scanning line is output.

  As described above, the motion detection unit 4 and the motion adaptive interpolation calculation unit 5 perform a video signal representing the value of a pixel located in the vicinity in the interpolation target field with respect to the video signal (Ib) of the interpolation target field, for example, Intra-field interpolation for generating an interpolation signal by an arithmetic process based on an adjacent scanning line, or a field temporally adjacent to an interpolation target field, for example, a field for generating an interpolation signal by an arithmetic process based on a temporally previous video signal Ic A motion adaptive interpolation processing means (first interpolation processing means) 45 for performing interpolating is configured.

  The telecine detection means (second interpolation processing means) 6 receives the current field signal (Ia) from the input terminal 100 and the 2-field delay signal (Ic) from the 1-field delay means 3. The telecine detecting means 6 detects a 2-3 pull-down field sequence from the difference information between one frame, detects whether or not the input video signal Ia is a telecine video signal, and detects a field sequence, A telecine detection signal tci indicating detection of the telecine video signal and a sequence phase signal tcf indicating the phase of the 2-3 pulldown field sequence are generated and output. The sequence phase signal tcf, which is the output of the telecine detection means 6, is sent to the telecine interpolation processing means 7 and the telecine detection signal tci is sent to the selection means 8 as a control signal. The detailed configuration of the telecine detection means 6 will be described later.

  The telecine interpolation processing means 7 receives the current field signal (Ia) from the input terminal 100 and the two-field delay signal Ic from the one-field delay means 3 and receives a sequence from the telecine detection means 6, particularly the telecine sequence detection means 21. Based on the phase signal tcf, depending on the phase of the field sequence, the signal in the field preceding the interpolation target field (Ib) (2 field delayed signal) (Ic) or the field after the interpolation target field (current field) ) Signal (Ia) is selected and output as an interpolated scanning line signal (telecine interpolation signal) It of the interpolation target field (Ib).

The selection unit 8 selects either the scanning line interpolation signal Im from the motion adaptive interpolation processing unit 45 or the telecine interpolation signal It from the telecine interpolation processing unit 7 according to the telecine detection signal tci output from the telecine detection unit 6. The interpolated scanning line signal I is sent to the terminal 102. The telecine detection signal tci is a signal indicating a detection result as to whether or not it is a telecine video signal. When the telecine detection signal tci indicates that it is a telecine video signal, the selection means 8 starts from the telecine interpolation processing means 7. When the telecine interpolation signal It is selected and the telecine detection signal tci indicates that it is not a telecine video signal, the signal Im from the motion adaptive interpolation processing means 45 is selected as a normal video signal.
Such selection is performed for each field.

  Then, the 1-field delay signal R from the 1-field delay means 2 is output to the terminal 101 as an actual scanning line signal, and the interpolation signal from the selection means 8 is output to the terminal 102 as the interpolation scanning line signal I. Then, the output from the terminals 101 and 102 is double-speed converted by the combining circuit 9 and is alternately switched and combined for each scanning line to obtain a sequentially scanned video signal PR.

  The above-described motion adaptive interpolation processing means 45, telecine interpolation processing means 7 and selection means 8 constitute a scanning line interpolation means 40.

The telecine detection unit 6 includes a correlation determination unit 30 and a telecine determination unit 20.
The correlation determining means 30 detects the correlation between the video signal of the target field (Ia) of the video signal and the field (Ic) one frame before the target field, and based on the strength of the correlation between frames, It is determined whether the field is a still image field or a moving image field.

  For example, the correlation determining means 30 is a pixel in which a difference value between each pixel in a predetermined area (region) in the target field of the video signal and a pixel at the same position in the field one frame before is larger than a predetermined value. The field of interest is a still image field (the amount of motion is below a predetermined value compared to the previous frame) or a moving image field (the amount of motion compared to the previous frame) Is greater than a predetermined value).

  The correlation determination unit 30 includes a pixel difference detection unit 11, a difference comparison unit 12, a predetermined area accumulation counter 13, and a field state determination unit 14.

The current field signal from the input terminal 100 and the 2-field delay signal from the 1-field delay means 3 are input to the pixel difference detection means 11 in the correlation determination means 30.
The pixel difference detecting means 11 calculates a difference value between each pixel of the video signal of the current field (video signal of the target field) and a pixel at the same position of the two-field delay signal (video signal one frame before the target field). The difference information (difference information between one frame) based on this difference value, for example, the absolute value of the difference is detected and output.

The difference comparison unit 12 receives the inter-frame difference information from the pixel difference detection unit 11, compares it with a predetermined threshold value, and outputs a comparison result. By this comparison, the presence or absence of motion in each pixel is determined, and the comparison result is used as the determination result of the presence or absence of motion.
In the comparison, for example, if the inter-frame difference information is equal to or less than a predetermined threshold, the pixel is determined as a still pixel, and if it is larger than the predetermined threshold, it is determined as a moving pixel, and as a result, a binarized value is output. The binarized value is, for example, a value “0” when it is determined as a still pixel, and a value “1” when it is determined as a moving pixel. The binarized value is not limited to this, and may be a value indicating the result of each pixel determined as a still pixel or a moving pixel by comparison.

The predetermined area accumulation counter 13 counts pixels determined to be moving with respect to the pixels in the predetermined area in the field based on the output from the difference comparison means 12. A binarized value as a moving picture element, an area signal te indicating a predetermined area in one field from the timing signal generating means 1, and a field pulse Vp are input. In the area indicated by the area signal te indicating the predetermined area in one field from the timing signal generating means 1 according to the value binarized as a still pixel or a moving pixel, the predetermined area accumulation counter 13 Count the number of pixels of the value that is a moving picture element. That is, the case where the value determined as a moving pixel from the difference comparison unit 12 is “1” is counted over the area of the area signal te, and is counted according to the field pulse Vp from the timing signal generation unit 1. The result (count value) is output as a value (number of moving pixels) corresponding to the number of moving pixels in one field.
In place of counting moving pixels, still pixels may be counted.

  Here, depending on the video signal, there is a case where it is sent as a movie material in which a black signal portion (no-image portion) exists in the upper and lower band portions called letter boxes, and the area signal te is, for example, this upper and lower in one field Is generated in the timing signal generating means 1 as a signal indicating a pixel period excluding the band portion and the image blanking period. Accordingly, the predetermined area accumulation counter 13 outputs the number of moving pixels in the area excluding the upper and lower band portions and the image blanking period as the number of moving pixels in one field, thereby preventing erroneous determination.

  As described above, instead of obtaining the difference for all pixels in one field in the pixel difference detecting means 11 and counting the moving pixels (pixels having a large difference) in the predetermined area by the predetermined area cumulative counter 13. In the pixel difference detecting means 11, a difference is obtained only for pixels in a predetermined area, and all moving pixels (pixels having a large difference) are obtained by using a cumulative counter that does not limit the area instead of the predetermined area cumulative counter 13. It is good to count.

  In addition, as described above, as the correlation determination unit 30, for each pixel in a predetermined area in the field of interest, a difference value between the pixel at the same position in the field before one frame is larger than a predetermined value. Instead of determining whether the field of interest is a still image field or a moving image field based on whether the number is equal to or greater than a predetermined number, for each pixel in the entire area of the field of interest, between the pixels at the same position in the field one frame before It is also possible to use one that determines whether the field of interest is a still image field or a moving image field based on whether the number of pixels having a difference value greater than a predetermined value is greater than or equal to a predetermined number.

  The field state determination unit 14 compares the number of moving pixels (counting result) in one field output from the predetermined area accumulation counter 13 with a predetermined value, and determines whether the field (target field) is a still image field or a moving image field. Determine if there is.

Here, the field state determination means 14 can set two predetermined values for comparison. For example, as shown in FIG. 2, the first comparison means 14a having a first predetermined value (threshold value) and And a second comparison unit 14b in which a second predetermined value (threshold value) is set, from which a first signal S1 and a second signal S2 indicating the comparison result are output. The first and second signals both have a value of “1” when indicating that they are still image fields, and have a value of “0” otherwise.
As will be described later, the first signal is used by the telecine sequence detection means 21, and the second signal is used by the sequence deviation detection means 22.
The first and second predetermined values may be different from each other, or may be the same value.

When the number of moving pixels input to the field state determination means 14 (that is, the number of moving pixels in a predetermined area of the field subjected to pixel counting) is equal to or less than the first predetermined value T1, the value of the signal S1 is “1”. ", And the result determined with reference to the first predetermined value T1 is" still image ", otherwise, the value of the signal S1 is" 0 "and the first predetermined value T1 is The result determined based on the reference is “moving image”.
Similarly, when the number of moving pixels (that is, the number of moving pixels in a predetermined area of the field subjected to pixel counting) input to the field state determination unit 14 is equal to or smaller than the second predetermined value T2, the value of S2 is “1” is obtained, and the result determined with reference to the second predetermined value T2 is “still image”. Otherwise, the value of S2 is “0” and the second predetermined value T2 It is indicated that the result of determination based on is “moving image”.
Accordingly, the signals S1 and S2 indicating the detection results based on the first and second predetermined values indicate whether the field is a still image or a moving image, and these signals S1 and S2 are output from the correlation determination unit 30 as telecine. It is output to the determination means 20.

  The telecine determination unit 20 receives the field signals S1 and S2 from the field state determination unit 14 and the field pulse Vp from the timing signal generation unit 1. Then, at the timing of the field pulse Vp, a 5-field sequence of the telecine video signal is detected, and whether the input video signal is a telecine video signal and the phase of the 2-3 pulldown field sequence (2-3 pulldown phase). And a telecine detection signal tci indicating the detection of the telecine video signal and a sequence phase signal tcf indicating the phase of the field sequence are output.

The telecine determination unit 20 includes a telecine sequence detection unit 21 and a sequence deviation detection unit 22.
The telecine sequence detection means 21 detects a telecine conversion field sequence in the video signal telecine-converted by the 2-3 pull-down method based on the signal S1 from the correlation determination means 30 for five fields, and the telecine conversion field sequence From the detection results, it is determined whether the sequentially input video signal is a telecine converted video signal (tci, tcf).
The sequence shift detection means 22 is based on the output signal S2 from the correlation determination means 30 and the detection results tci and tcf of the telecine sequence detection means 21, and in the telecine conversion field sequence detected by the telecine sequence detection means 21, This is detected when a field different from the phase field to be obtained is a still picture field.
As will be described later, in a video signal subjected to telecine conversion, one sequence is formed by five fields, and a field of a specific phase (order in the sequence) among them is a still image field. That is, a field that becomes a still image appears once every five fields. Such a phase is referred to herein as a “specific phase”. In the present invention, when it is detected that a field other than the field of the specific phase is a still picture field, the sequence shift is detected based on this.
The telecine sequence detection means 21 also uses the output of the sequence deviation detection means 22 to determine whether or not the video signal has been telecine converted.
As described in detail below, based on the detection result from the telecine sequence detection means 21, the process for generating the interpolated scanning line signal in the scanning line interpolation means 40 is switched.

FIG. 3 is a block diagram showing a configuration of an example of the telecine determination unit 20.
Although the field pulse Vp is not shown in FIG. 3, the operation in each block is performed in units of fields and operates based on the timing of the field pulse Vp from the timing signal generating means 1.

  The telecine sequence detecting means 21 in the telecine determining means 20 shown in FIG. 3 includes a moving picture field counter 211, a telecine sequence phase detecting means 212, a moving picture field count value comparing means 213, a still picture field continuous detecting means 214, and a telecine. Detection signal generation means 215.

The moving picture field counter 211 detects the number of consecutive fields determined as moving picture fields by the correlation determining means 30.
The moving image field counter 211 receives the signal S1 from the field state determination unit 14. The moving image field counter 211 counts (increments the count value by 1) every time the signal S1 is reset in a field indicating a still image field and indicates that the signal S1 is a moving image field, and at the timing of the field pulse Vp. The count value mf up to the previous field is output as the continuous number of moving image fields.

The telecine sequence phase detection unit 212 detects a telecine conversion field sequence in the telecine-converted video signal based on the signal S1 from the correlation determination unit 30 and the output mf from the moving image field counter 211.
The telecine sequence phase detection means 212 receives the signal S1 from the field state determination means 14 and the continuous number mf of moving picture fields from the moving picture field counter 211. The telecine sequence phase detecting means 212 repeatedly counts from 0 to 4 every 5 fields based on the field pulse Vp, and counts the repetition period based on the signal S1 and the continuous number mf of moving image fields up to the previous field. Reset. That is, when the signal S1 is “1” (indicating that it is a still image field) and the continuous number mf of moving image fields up to the previous field is “4”, a field of a specific phase in the telecine video signal is a still image. Assuming that it is a field, the count value is reset to “0”. Then, the count value is output as the sequence phase signal tcf. This signal tcf is repeated in units of 5 fields, and is reset in synchronism with a still image field having a specific phase once every 5 fields. Therefore, this signal tcf indicates the phase of a repeated sequence of 5 fields. And sent to the sequence deviation detecting means 22.

For example, as shown in FIG. 4, the telecine sequence phase detection unit 212 uses the signal S1 from the correlation determination unit 30 and the output mf from the video field counter 211 to display a still image after four consecutive video fields. A specific phase field detection unit 212a for detecting the timing of a field to be a field, and a phase signal generation unit 212b for generating a signal tcf that repeats in a five-field cycle in synchronization with the still picture field detected by the specific phase field detection unit 212a; The signal output from the phase signal generating means 212b is output as a signal tcf indicating the phase of the field sequence, which is the 2-3 pulldown conversion phase.
The phase signal generating unit 212b includes, for example, a counter that counts the field pulse Vp and is reset by the output of the specific phase field detecting unit 212a.

  The moving picture field count value comparison means 213 receives the moving picture field continuous number mf from the moving picture field counter 211, compares the moving picture field continuous number mf with a predetermined value, and detects the case where the continuous number is equal to or greater than the predetermined value. The detection result is output. In the telecine video signal, there are 5 fields that are the same as 2 fields before, so that the continuous number mf of moving image fields from the moving image field counter 211 does not exceed 5 fields, and the predetermined value used for detection is “ 5 ”, a detection result (213OUT) that becomes“ 1 ”, for example, when 5 fields or more are continuous is output and sent to the telecine detection signal generation means 215.

The still image field continuous detection unit 214 generates a signal indicating this when a field determined to be a still image by the correlation determination unit 30 is continuously generated a predetermined number of times or more. It has a field counter 214a and a still image field count value comparison means 214b.
The still image field counter 214a receives the signal S1 from the field state determination unit 14 of the correlation determination unit 30, counts the field (still image field) where the signal S1 is “1”, and the count value is the still image field. When the count value comparing means 214b compares with the predetermined value and determines that the value is equal to or larger than the predetermined value, the signal 214OUT becomes “1”, thereby detecting the case where the still image field continues for the predetermined number of fields or more.
Therefore, first, the count value of the still image feed counter 214a is reset to zero when the signal S1 indicates the moving image field, and the count value is increased by 1 when the signal S1 indicates the still image field.
Then, the still image field count value comparison means 214b compares the count value of the still image field with a predetermined value, and when the count value (continuous number) exceeds the predetermined value, the output 214OUT is set to “1” to detect telecine. The signal is sent to the signal generation means 215. The predetermined value used for the comparison by the still image field count value comparison means 214b is set to “5”, for example, in consideration of a five-field repetitive sequence of the telecine video signal. That is, the output 214OUT becomes “1” when 5 fields or more are continued.

The telecine detection signal generation means 215 outputs the field sequence repetition number, the count value mf from the moving picture field counter 211, the output 214OUT from the still picture field continuous detection means 214, and the output det_BE from the sequence deviation detection means 22. Based on the signal S1 from the correlation determination means 30 and the count value mf from the moving picture field counter 211, the number of repetitions of the detected telecine conversion field sequence is obtained, and the video signals sequentially input are telecine converted. Whether or not is detected.
The output tci of the telecine detection signal generation means 215 is output as a signal indicating whether or not it is a video signal subjected to telecine conversion, and the output tcf of the telecine sequence detection means 212 is output as a signal indicating the phase of the telecine conversion field sequence. The

  As shown in FIG. 5, for example, the telecine detection signal generation unit 215 outputs 213OUT from the moving image field count value comparison unit 213, output 214OUT from the still image field continuous detection unit 214, and sequence deviation detection unit 22. Based on the output det_BE, reset condition determining means 215a for determining a reset condition, reset by the output of the reset condition determining means 215a, and based on the signal S1 from the correlation determining means 30 and the count value mf from the moving picture field counter 211. Then, after the moving image field continues for four fields, the field sequence continuation number detecting means 215b for obtaining the number of continuations of the field sequence that becomes the still picture field, and the continuity number (counting) obtained by the field sequence continuity number detecting means 215b Value) fsc is compared with a predetermined value, and a field sequence continuous number comparing means 215c is provided, and a signal tci indicating whether or not the output signal of the field sequence continuous number comparing means 215c is the above-mentioned telecine-converted video signal. Output as.

The field sequence continuous number detecting means 215b indicates the count value when the output signal S1 from the correlation determining means 30 indicates a still picture field and the output mf from the moving picture field counter 211 indicates 4. For example, as shown in FIG. 5, the moving image field continuous number comparison means 215e, an AND gate 215f, and a field sequence counter 215g are provided.
The moving picture field continuous number comparing means 215e determines a field where mf is “4”, and outputs “1” when mf is “4”.
The AND gate 215f outputs "1" when both the output of the moving picture field continuous number comparing means 215e and the signal S1 are "1".
The field sequence counter 215g is reset by the output of the reset condition determination unit 215a and counts the output of the AND gate 215f.

  When the output 213OUT from the moving picture field count value comparing means 213 indicates that the number of continuous moving picture fields is a predetermined value, for example, 5 or more, the reset condition determining means 215a When the output 214OUT indicates that the number of consecutive still image fields is a predetermined value, for example, 5 or more, the output det_BE from the sequence deviation detection means 22 indicates that a sequence deviation has occurred. When any of the above conditions is satisfied, the field sequence counter 215g of the field sequence continuous number detecting means 215b is reset.

  First, detection of whether or not the video signal is a telecine video signal and generation of the telecine detection signal tci will be described. The telecine detection signal generating means 215 counts the case where a sequence of 5 fields, which becomes a still image field after 4 consecutive moving image fields, is detected based on the signal S1 and the continuous number mf of moving image fields up to the previous field. That is, if the signal S1 is “1” (indicating that it is a still image field) and the number of continuous video fields up to the previous field is mf = 4, a sequence of 5 fields of the telecine video signal is detected as 5 The field sequence counter 215g counts the number of repetitions of the field sequence. When the count value exceeds a predetermined value, that is, when a sequence of five fields of a telecine video signal continues for a predetermined number of times, this is detected by the field sequence continuous number comparison means 215c, and the input video signal is When it is detected as a telecine video signal, a telecine detection signal tci is generated and output. For example, when the telecine detection signal tci is detected as a telecine video signal, the telecine detection signal tci is output as “1”, and when it is other than the telecine video signal or until it is detected (confirmed) as a telecine video signal. If it is not a telecine video signal, for example, during detection or when detection is canceled, “0” is set. Note that the value of the telecine detection signal is not limited to this, and it is only necessary to be able to discriminate between the case where the telecine video signal is detected and the case where the other signal is used.

  When the telecine detection signal tci is canceled by the telecine detection signal generation means 215 (that is, when the input video signal is not a telecine video signal), it is detected that a moving image field continues for a predetermined field (5 fields) or more. When a still picture field is detected to continue for a predetermined field (5 fields) or more, and during the detection of a telecine video signal, a 2-3 pulldown field sequence is changed by editing or the like. When the phase is shifted, the count value fsc in the field sequence counter 215g that counts the number of consecutive 5 field sequences is reset to zero, and the telecine detection signal tci is set to invalid "0". That is, when any detection result is sent based on the output of the moving image field count value comparison unit 213, the output of the still image field continuous detection unit 214, and the output det_BE from the sequence deviation detection unit 22, Is reset to zero, and the telecine detection signal tci is set to “0”, so that the telecine detection signal tci is canceled and it is detected that it is not a telecine video signal. After the cancellation, when the above five-field sequence is detected again, counting by the field sequence continuous number detecting means 215b and comparison by the field sequence continuous number comparing means 215c are performed again, and the telecine detection signal tci is determined based on the comparison result. The operation in which becomes “1” is repeated.

  The sequence deviation detection means 22 is configured as shown in FIG. 6 and is different from a field of a specific phase once in the middle of a sequence of 5 fields during detection of a telecine video signal (tci = 1). Is a still picture field, and a 2-3 pull-down phase shift of the field sequence is detected. FIG. 6 shows an exemplary configuration of the sequence deviation detection means 22. 6 includes a still image field counter 221 and a count value comparison unit 222, and an output of the count value comparison unit 222 is output as an output det_BE of the sequence deviation detection unit 22.

  The still image field counter 221 includes a signal S2 indicating whether or not the image is a still image from the field state determination unit 14, a telecine detection signal tci from the telecine sequence detection unit 21, and a sequence phase signal tcf indicating a 2-3 pull-down phase. Entered. When the telecine detection signal tci from the telecine sequence detection means 21 indicates that the input video signal is a telecine video signal (that is, when tci = 1), the still picture field counter 221 receives the signal from the field state determination means 14. A field indicating that the signal S2 is a still picture field is counted. However, the still picture field is counted every cycle of the sequence phase signal tcf indicating the 2-3 pull-down phase, and the count value is set to “0” in the still picture field (tcf = 4) having a specific phase once every five fields. Reset to “”. Therefore, the count value is increased by 1 every time the signal S2 becomes “1” in a field different from the field of the specific phase once in 5 fields in the telecine video signal, and thereby the signal S2 becomes “1”. And the count value is output. The count value 221OUT in the still image field counter 221 is output to the count value comparison means 222.

  The count value comparison means 222 compares the count value 221OUT from the still image field counter 221 with a predetermined value, and if the count value is equal to or greater than the predetermined value, a signal indicating detection of a still image field in the middle of a five-field sequence, for example, “ Output as 1 ″. In the case of a telecine video signal, a field with a specific phase once becomes a still image field every 5 fields, so if it is detected that a field different from the field with the specific phase in the middle of the sequence is a still image field, Therefore, it can be determined that there is a 2-3 pull-down phase shift. Therefore, this comparison result is output as a signal det_BE indicating a 2-3 pull-down phase shift. The predetermined value used in the above comparison is, for example, “1”.

  The output det_BE from the comparison means 22 in the sequence deviation detection means 22 generates a telecine detection signal as a signal indicating a 2-3 pull-down phase deviation in order to cancel the telecine detection signal tci indicating that it is a telecine video signal. Sent to means 215.

  As described above, the telecine detection signal generation means 215 generates the telecine detection signal tci, and the output det_BE from the comparison means 22 in the sequence deviation detection means 22 has a 2-3 pulldown phase in the middle of the telecine video signal. Even when a deviation is detected, the count value fsc of the field sequence counter 215g is reset to zero, and the telecine detection signal tci = 0 is set. Therefore, when the 2-3 shift of the pull-down phase is detected in the middle of the telecine video signal by the sequence shift detection means 22, the telecine detection signal tci becomes tci = “0”, the telecine detection signal is canceled and the telecine is detected. 1 indicates that the signal is not a video signal, and the selection unit 8 in FIG. 1 selects the signal from the motion adaptive interpolation processing unit 45 as a normal video signal.

  Next, the operation of the telecine detection means 6 for detecting a telecine video signal will be described, assuming that a telecine video signal having no editing point or the like and having a 2-3 pull-down phase sent regularly is input from the input terminal 100. .

  FIG. 7 is a diagram showing a relationship between a telecine video signal in the telecine conversion by the 2-3 pull-down method and a video signal before conversion. A first frame A of a signal such as a movie film corresponds to the first and second fields At and Ab of the video signal, and a second frame B corresponds to Bt, Bb and Bt of the third to fifth fields of the video signal. The third frame C is converted to correspond to the sixth and seventh fields Cb and Ct of the video signal, and the fourth frame D is converted to correspond to Db, Dt and Db of the eighth to tenth fields of the video signal. The subscript t in the figure represents a top field that is a relatively upper line field, and the subscript b represents a bottom field that is a lower line field.

  From FIG. 7, once every 5 fields, 2 fields so that the third field Bt and the fifth field Bt are fields having the same contents, and the eighth field Db and the tenth field Db are fields having the same contents. Repeat the same field as before. Therefore, the inter-frame difference information obtained by the pixel difference detecting means 11 in the telecine detecting means 6 is zero or a predetermined value (a value considering noise close to zero) in the detection in the fifth field and the tenth field. It becomes as follows. This 1-frame difference information relationship is repeated once every 5 fields.

FIG. 8 is a diagram showing signals appearing in each part of the telecine detection means 6 of FIG. 1 when the telecine video signal shown in FIG. 7 is input. It is a figure for demonstrating the detection operation of the telecine detection means. (B), (c) and (d) in FIG. 8 are the interlaced video signal Ia input from the input terminal 100, the 1-field delay signal Ib from the 1-field delay means 2, and 2 by the 1-field delay means 3. The field delay signal Ic is shown. The pixel difference detection means 11 detects the inter-frame difference information between the input signal and the two-field delay signal, and the difference comparison means 12 binarizes the inter-frame difference information. The number of moving pixels in the field is obtained by counting the number of pixels that become moving pixels in a predetermined area. Since the field state determination means 14 compares the number of moving pixels in one field with a predetermined value and detects whether or not the field is in the still image field state, the signal S1 is shown in FIG. 8 (e). As shown in the figure, at the field position next to the fifth field and the tenth field, a value indicating a still picture field ("1" in the figure) is obtained, and a complete still picture field is detected once in five fields. In FIG. 8E, “0” is a value indicating that the signal S1 is a moving image field.

  The moving image field counter 211 in the telecine sequence detection means 21 is reset to “0” when the signal S1 is “1” (indicating that it is a still image field), and when the signal S1 is “0” (moving image). When the field is indicated), the count value is incremented by 1, thereby outputting the continuous number mf of moving image fields up to the previous field (FIG. 8 (f)). The telecine sequence phase detection means 212 repeatedly counts from 0 to 4 every 5 fields, but the signal S1 is “1” (indicating that it is a still image field), and the continuous number mf of the moving image field is 4. At this time, the count value is reset to “0” on the assumption that the still picture field is detected once in the sequence of the telecine video signal and 5 fields.

  FIG. 8 (g) shows an example of the sequence phase signal tcf, where the signal S1 is “1” and the continuous number mf of the moving image field is “4” (in the next field), and the count value is 0. This is an example. From the figure, the signal repeats in units of 5 fields. In this case, the phase of 2-3 pull-down is as follows: when the sequence phase signal tcf = 0, 2, the input current field (Ia) and the 1-field delay signal Ib In the sequence phase signal tcf = 3, the input current field Ia and the 2-field delay signal Ic have the same contents, and when tcf = 1, 4, the 2-field delay signal Ic and the 1-field delay signal Ib It is a field from the same frame. Therefore, in the telecine interpolation processing means 7, the video signal Ia in the input current field is obtained when the sequence phase signal tcf = 0, 2 in accordance with the sequence phase signal tcf, and the input current field Ia or 2 in the sequence phase signal tcf = 3. When the field delay signal Ic is tcf = 1, 4, the 2-field delay signal Ic is selected as the telecine interpolation signal It (for the interpolation target field (Ib)) and output. In the figure, the output timing of the count value “0” is the position of the next field, but it can also be output at the timing immediately after the start of the field with the signal S1 = “1” and the continuous number of moving image fields mf = 4. The timing is based on the field pulse Vp and is not limited to the above example.

  The telecine detection signal generation means 215 counts the case where the signal S1 in FIG. 8 is “1” and the continuous number mf of the moving image field is 4, the count value becomes equal to or greater than a predetermined value, and continues for a predetermined number of times. If the sequence is detected, it is assumed that the detection of the telecine video signal is valid, and the telecine detection signal tci = 1 is set. The telecine detection signal tci in FIG. 8 (i) shows a case where the sequence count is, for example, two times or more and it is detected that the input video signal is a telecine video signal. Therefore, when a sequence in which a field of a specific phase once every 5 fields becomes a still image field is continuously detected twice or more, and the telecine detection signal tci = 1, it is detected that the input video signal is a telecine video signal. Thus, the selection means 8 selects the telecine interpolation signal It from the telecine interpolation processing means 7.

  Note that when switching from a telecine video signal to a video signal for normal television broadcasting or the like, the interframe difference information detected by the pixel difference detection means 11 becomes a large value and is output from the field state determination means 14. In the signal S1, the value “0” indicating that it is a moving image field is continuous. Therefore, when the number mf of video field continuations reaches 5 fields or more, the detection result 213OUT = "1" indicating the continuation of 5 fields or more is sent from the video field count value comparison unit 213 to the telecine detection signal generation unit 215. Sent.

  The telecine detection signal generation means 215 resets the count value of the field sequence counter 215g to zero when the signal from the moving picture field count value comparison means 213 becomes “1”, and sets the telecine detection signal tci = 0. Therefore, when the video signal is switched from the telecine video signal to a video signal in which five or more moving image fields continue, the telecine detection signal tci is canceled to 0, and the selection means 8 is a normal video signal that is not a telecine video signal. As a result, the signal from the motion adaptive interpolation processing means 45 is selected.

  In addition, when the telecine video signal is switched to a video signal in which a still image field is continuous, the inter-frame difference information detected by the pixel difference detection unit 11 becomes a small value and is output from the field state determination unit 14. In the signal S1, the value “1” indicating that it is a still image field is continuous. Therefore, the continuous number mf of the moving image field is always 0, and the still image field continuous detection unit 214 counts the still image field, so that 214OUT = “1” is detected when the continuous number becomes 5 fields or more. The result is sent to the telecine detection signal generation means 215.

  In the telecine detection signal generation means 215, when the signal 214OUT from the still image field continuous detection means 214 becomes "1", the count value (continuous number) of the field sequence continuous number detection means 215b is reset to zero to detect the telecine. The signal tci is set to “0”. Therefore, when the telecine video signal is switched to a video signal in which a still image field continues for five or more fields, the telecine detection signal tci is canceled to “0”, and the selection unit 8 causes the normal video that is not a telecine video signal to be displayed. The signal from the motion adaptive interpolation processing means 45 is selected as a signal.

  Next, in the middle of the telecine video signal sequence as shown in FIG. 9, a telecine video signal in which the regularity of the field sequence is disturbed by editing for broadcasting, insertion of CM or telop, etc. is input from the input terminal 100. The operation of the telecine detection means 6 will be described.

  FIG. 9 shows an example where there is an edit point in the sixth field in the middle of the telecine video signal and the sequence regularity is disturbed in the sixth field Cb. In FIG. 9, the third field Bt and the fifth field Bt are fields having the same contents, and the seventh field Dt and the ninth field Dt are fields having the same contents.

  FIG. 10 is a diagram for explaining the detection operation of the telecine detection means 6 when the telecine video signal as shown in FIG. 9 is input. In the drawing, the basic operation until the values of the signals S1 and S2 are determined by the field state determination means 14 shown in (b) to (d) and (e) is the same as that in the above-described detection of the telecine video signal. Therefore, although detailed description thereof is omitted, the signals S1 and S2 indicate that they are still image fields at the field positions next to the fifth field and the ninth field, as shown in FIG. The value is “1”. However, it is assumed that the thresholds for determining the values of the signals S1 and S2 are the same.

  The moving picture field counter 211 outputs the continuous number mf of moving picture fields up to the previous field (FIG. 10 (f)). The telecine sequence phase detection means 212 repeatedly counts from 0 to 4 every 5 fields to obtain the sequence phase signal tcf, which indicates that the signal S1 is a still image field, and the continuous number of moving image fields mf = When 4, the count value is reset to “0”.

  In the example shown in FIG. 9, the regularity of the sequence is disturbed in the sixth field, so that the continuous number mf of the moving image field is “1” even though the signal S1 is “1” in the next field after the ninth field. 3 ", which reflects the phase shift of 2-3 pulldown.

  Therefore, in order to determine the 2-3 pull-down phase shift, the sequence shift detection means 22 detects that a field different from the specific phase field once every 5 fields is a still image field.

  In the still picture field counter 221 of the sequence deviation detecting means 22, when the telecine detection signal tci = 1, the count value 221OUT is reset to “0” every sequence phase signal tcf = 4. When the 2-3 pull-down phase of the telecine video signal is sent regularly, the field with the sequence phase signal tcf = 4 is a field (signal S1 is “1”) that becomes a still image after 4 moving image fields continue. Field), that is, a still image field having a specific phase once every 5 fields. Therefore, when a still image field having a specific phase is detected once in 5 fields, the count value 221OUT is reset to “0”. Is done. The signal S2 (FIG. 10 (e)) from the field state determination means 14 is a still image until it is reset to “0” (until the field of the specific phase of the next field sequence). Count the fields that indicate. Therefore, the still picture field is counted in the still picture field counter 221 when the ninth field still picture field is detected.

  Therefore, a field in which the signal S2 is “1” in a field different from the field of one specific phase in a 5-field sequence of the telecine video signal is detected, and the number of the detected fields is determined as the still image field counter 221. And the count value 221OUT is output. The count value comparison means 222 compares the count value 221OUT from the still picture field counter 221 with a predetermined value, for example, “1”, and when it is equal to or larger than the predetermined value, sets the signal det_BE = 1 indicating a 2-3 pull-down phase shift. Output (FIG. 10 (i)).

  The output det_BE from the comparison means 22 in the sequence deviation detection means 22 is sent to the telecine detection signal generation means 215. When the telecine detection signal generation means 215 becomes det_BE = 1, the field sequence continuous number detection means 215b. Is reset to zero, and the telecine detection signal tci is set to “0”. Therefore, when a 2-3 pull-down phase shift is detected in the middle of the telecine video signal, the telecine detection signal tci is canceled to “0”, and the selection unit 8 selects the signal from the motion adaptive interpolation processing unit 45. It will be. Even after the cancellation, the telecine sequence detection means 21 performs the operation of detecting the telecine video signal. If a five-field sequence is regularly input and accurately detected, the telecine detection signal tci = 1 again. Thus, interpolation corresponding to the telecine video signal is performed.

  Therefore, when it is detected that the regularity of the field sequence is disturbed in the middle of the telecine video signal, the IP conversion operation based on the detection of the telecine video signal can be temporarily canceled, and the wrong phase The signal from the motion adaptive interpolation processing means 45 is selected until the telecine conversion field sequence is accurately detected without generating the interpolation signal of the telecine video signal.

  As described above, according to the first embodiment, information indicating whether the field is a still image or a moving image is obtained from the inter-frame difference information, and the telecine sequence detection means 21 detects the sequence of five fields of the telecine video signal. The sequence deviation detection means 22 detects that when a field different from a specific phase field once every 5 fields is a still picture field, and detects the telecine detection signal generated by the telecine detection signal generation means 215. Release tci temporarily. Therefore, even when a 2-3 pull-down phase shift occurs in the telecine video signal, the image quality such as the image shift by stably detecting the telecine video signal and generating the scanning line interpolation signal by the erroneous field signal is obtained. It is possible to prevent deterioration and perform IP conversion of a good telecine video signal.

  In the predetermined area accumulation counter 13 in the first embodiment, the number of moving pixels in a predetermined area of one field is obtained, and the field state determination means 14 compares the number of moving pixels with a predetermined value to determine whether the field is a still image or a moving image. In the predetermined area cumulative counter 13, the number of still pixels is obtained, and the field state determination means 14 compares the number of still pixels with a predetermined value to determine whether the field is a still image or a moving image. The same effects as those of the first embodiment are obtained.

  In the first embodiment, the telecine detection means 6 is provided with the pixel difference detection means 11 to detect difference information between one frame. However, the motion detection means 4 also detects the current field signal and the two field delay. It is also possible to obtain difference information between signals and to detect a telecine video signal based on one frame difference information in the motion detection means 4. Further, the pixel difference detecting means 11 takes the absolute value of the difference between one frame and outputs it as difference information. However, the pixel difference detecting means 11 performs nonlinear conversion such as multiplying the absolute value of the difference by a predetermined value and subtracting the predetermined value. The converted value may be used as 1-frame difference information, and it is only necessary to determine whether the field is a still image or a moving image from the obtained 1-frame difference information.

  Further, the still image field continuation detecting unit 214 in the first embodiment counts the still image field by receiving the signal S1 indicating whether or not it is a still image from the field state determining unit 14, but the field state determining unit 14 A third predetermined value that is different from the first and second predetermined values is provided, and the output of the predetermined area accumulation counter 13 is compared with the third predetermined value to determine whether or not the image is a still image. It is also possible to output the third signal S3, and to count the field indicated by the third signal S3 as a still image by the still image field continuous detection means 214. In this case, the third predetermined value can be set to a value that takes into account when a still pixel such as a caption is inserted.

  The telecine sequence phase detection means 212 in the first embodiment repeats counting from 0 to 4 every 5 fields to indicate that the signal S1 is a still image field, and the number of continuous moving image fields mf = 4. Although it is sometimes reset to “0”, the repetition value and the reset value are not limited to this. This signal repeats every 5 fields, and indicates that the signal S1 is a still image field, and when the number of continuous moving image fields mf = 4, that is, synchronizes with a still image field having a specific phase once every 5 fields. Then, it only needs to be reset to a predetermined value, and the same effect as in the first embodiment can be obtained.

  Further, the IP conversion for the normal video signal in the first embodiment has been described so as to perform the motion adaptive interpolation process. However, an interpolation signal by only the intra-field interpolation process may be used, and the same as in the first embodiment. The effect of.

  Furthermore, although the hardware configuration is described in the first embodiment, the configuration is not limited to the hardware configuration, and may be realized by software processing under program control.

Embodiment 2. FIG.
FIG. 11 is a block diagram showing a configuration of another example of the sequence deviation detecting means 22 in FIG.

  In FIG. 11, the sequence deviation detection unit 22 according to the second embodiment includes a still image field counter 221, a count value comparison unit 222, a delay unit 223, and a logic gate 224. Note that the still image field counter 221 and the count value comparison means 222 are the same as those in FIG. 6 of the first embodiment, and a detailed description thereof will be omitted.

  The still picture field counter 221 includes a signal S2 indicating whether the field is a still picture from the field state determination means 14, a telecine detection signal tci from the telecine sequence detection means 21, and a sequence phase signal tcf indicating 2-3 pulldown phase. When the telecine detection signal tci = 1, the field in which the signal S2 is “1” in the sequence of 5 fields indicated by the sequence phase signal tcf is counted. The count value comparison means 222 compares the count value from the still image field counter 221 with a predetermined value, and if it is equal to or larger than the predetermined value, a signal indicating that a still image field in the sequence of five fields has been detected, for example, “ Output as 1 ″.

  The delay means 223 delays the comparison result by the comparison means 22 by one field and outputs the comparison result in the previous field.

  The logic gate 224 receives the signal S2 from the field state determination unit 14 and the output from the delay unit 223, the signal S2 is “0” (indicating that it is a moving image field), and the output 223OUT of the delay unit 223. Is “1” (that is, when the comparison result of the previous field is “1”, that is, when a still image field in the middle of the sequence is detected in the previous field), a signal det_BE indicating a 2-3 pull-down phase shift = "1" is output.

  The count value comparison means 222, the delay means 223 and the logic gate 224 indicate that the signal S2 from the correlation determination means 30 is a moving image field, and the count value of the still image field counter 221 one field before The determination means (deviation presence / absence determination means) 225 is configured to generate a signal det_BE indicating whether or not the value indicates that it is greater than or equal to a predetermined value, and the output of the determination means 225 is output from the sequence deviation detection means 22. Is output as

  Accordingly, when the count value comparison unit 222, the determination unit 225 including the delay unit 223 and the logic gate is a still image field when a field different from the field of the specific phase once every five fields is a still image field. If it is detected and the next field is a moving image field, it is determined that the field has been changed, the regularity of the sequence is disturbed, and a 2-3 pull-down phase shift has occurred. In the field sequence of the telecine conversion, this determination condition is that the next field of the still image field is a moving image field, and if the next field of the still image field is a still image field, the direction before and after the field used for interpolation ( Even if there is an error in the field used for interpolation (whether it is in front or behind), it is considered that there will be no image shift. However, if a video field follows a still image field, the direction of the field used for interpolation This takes into account the fact that image misalignment may occur depending on (forward or backward).

  The output det_BE from the logic gate 224 of the determination means 225 in the sequence deviation detection means 22 is sent to the telecine detection signal generation means 215 in order to cancel the telecine detection signal as a signal indicating a 2-3 pull-down phase deviation. It is done.

  In the telecine detection signal generation means 215, the operation of resetting the count value of the field sequence continuation number detection means 215b and setting the telecine detection signal tc to “0” by the output det_BE from the sequence deviation detection means 22 is described in the first embodiment. It is the same as the case of. Therefore, when a 2-3 pull-down phase shift is detected in the middle of the telecine video signal by the sequence shift detection means 22, the telecine detection signal tci is canceled to "0", and the selection means 8 in FIG. The signal from the motion adaptive interpolation processing means 45 is selected as a normal video signal that is not a video signal.

  Next, an operation when a telecine video signal whose sequence regularity is disturbed by editing for broadcast, insertion of CM or telop, etc., is input from the input terminal 100 in the middle of the telecine video signal. The operations in the blocks other than the sequence deviation detecting means 22 such as the determination of whether the field is a still image or a moving image and the operation in the telecine sequence detecting means 21 are the same as those in the first embodiment. The description will be omitted, and the operation in the sequence deviation detecting means 22 will be mainly described.

  First, a case where a telecine video signal having a disordered sequence regularity in some fields as shown in FIG. 9 is input will be described. FIG. 12 is a diagram showing signals appearing in each part of the telecine detection means 6 of the second embodiment when the telecine video signal shown in FIG. 9 is input. In the figure, the operation other than the sequence deviation detection means 22 is the same as the operation in FIG. 10 described in the first embodiment.

  When the telecine detection signal tci = 1, the still image field counter 221 resets the count value 211OUT to “0” every time the sequence phase signal tcf = 4, and then resets the signal to “0”. A field in which S2 (FIG. 12E) is “1” (indicating that it is a still image field) is counted. As a result, a field that is different from a field of a specific phase once in a 5-field sequence in a telecine video signal and in which the signal S2 is “1” is counted (counted), and the count value is output. . The count value comparison unit 222 compares the count value 221OUT from the still image field counter 221 with a predetermined value, for example, “1”, and outputs a comparison result “1” when the value is equal to or larger than the predetermined value. The operations up to here are the same as those in the first embodiment.

  The output from the count value comparison means 222 is delayed by one field by the delay means 223, and the comparison result 223OUT in the previous field is obtained (FIG. 12 (i)). When the signal S2 is “0” and the output 223OUT from the delay means 223 is “1”, the logic gate 224 outputs the signal det_BE = “1” indicating a 2-3 pull-down phase shift. In the case shown in FIG. 12, the signal det_BE = "1" is output at the position of the eleventh field where the output 223OUT of the delay means 223 is "1".

  Next, as shown in FIG. 13A, the sequence deviation detecting means 22 when a telecine video signal whose sequence regularity is disturbed by inserting a still image in the middle of the telecine video signal is input. Will be described.

  FIG. 13 shows a case where edit points exist in the sixth to ninth fields Cb, Ct, Cb, and Ct in the middle of the telecine video signal, four fields are inserted as still images, and the regularity of the sequence is disturbed. It is a figure which shows the signal which appears in each part of the telecine detection means 6 of the form 2 of. In FIG. 13, the inter-frame difference information is zero or less than a predetermined value (a value considering noise close to zero) by detection in the fifth field, the eighth field, the ninth field, and the fourteenth field. .

  As shown in FIG. 13E, the signals S1 and S2 are values “1” indicating that they are still image fields at the field positions next to the fifth field, the eighth field, the ninth field, and the fourteenth field. It becomes.

  The moving picture field counter 211 outputs the continuous number mf of moving picture fields up to the previous field (FIG. 13 (f)). The telecine sequence phase detection means 212 repeatedly counts from 0 to 4 every 5 fields to obtain the sequence phase signal tcf, indicating that the signal S1 is “1” and is a still image field, When the continuous number mf = 4, the count value is reset to “0”. Since still images are inserted in the 6th to 9th fields, even if the signal S1 is “1” in the next field after the 8th and 9th fields to indicate that it is a still image field, The number mf is mf = 2 or 3, which reflects the 2-3 pull-down phase shift.

  In the case of the telecine detection signal tci = 1, the still image field counter 221 first resets the count value 221OUT to “0” every time the sequence phase signal tcf = 4, and then until it is reset to “0”. , The field where the signal S2 (FIG. 13 (e)) is “1” is counted. Therefore, the still picture field counter 221 counts when the still picture fields of the eighth and ninth fields are detected (S2 = 1). The count value comparison unit 222 compares the count value from the still image field counter 221 with a predetermined value, for example, “1”, and outputs a comparison result “1” when the count value is equal to or greater than the predetermined value, and passes through the delay unit 223. A comparison result 223OUT (FIG. 13 (i)) in the previous field delayed by one field is obtained.

  When the signal S2 is “0” and the output 223OUT of the delay unit 223 is “1”, the logic gate 224 outputs the signal det_BE = “1” indicating a 2-3 pull-down phase shift. (FIG. 13 (j)).

  The output det_BE of the logic gate 224 is sent to the telecine detection signal generation means 215. When the telecine detection signal generation means 215 becomes det_BE = 1, the count value fsc of the field sequence counter 215g is reset to zero, and the telecine detection signal generation means 215 The detection signal tci is set to tci = 0 (FIG. 13 (l)). Therefore, when a 2-3 pull-down phase shift is detected in the middle of the telecine video signal, the telecine detection signal tci = 0 is canceled and the selection means 8 selects the signal from the motion adaptive interpolation processing means 45. Become.

  Therefore, when it is detected that the regularity of the field sequence is disturbed in the middle of the telecine video signal, the IP conversion operation based on the detection of the telecine video signal can be temporarily canceled, and the wrong phase The signal from the motion adaptive interpolation processing means 45 is selected until the phase of the field sequence is accurately detected without generating the interpolation signal of the telecine video signal.

  As described above, according to the second embodiment, it is determined whether the field is a still image or a moving image from the inter-frame difference information, the telecine sequence detection unit 21 detects the telecine video signal, and the sequence deviation detection unit 22 Then, when a field different from a specific phase field once every five fields is a still image field, this is detected. When the moving image field is detected next to the detected still image field, the field sequence is changed, and it is determined that there is a 2-3 pull-down phase shift, and the telecine detection signal generation unit 215 detects the telecine detection signal. Release tci temporarily. Therefore, even when a 2-3 pull-down phase shift occurs in the telecine video signal, the image quality such as the image shift by stably detecting the telecine video signal and generating the scanning line interpolation signal by the erroneous field signal is obtained. Deterioration can be prevented and IP conversion of a good telecine video signal can be performed. In addition, when the field next to the still image field is a moving image field, a phase shift is detected, so that the telecine detection signal tci can be canceled more reliably and the telecine video signal can be detected stably.

  In the second embodiment, the hardware configuration is described. However, the configuration is not limited to the hardware configuration, and may be realized by software processing under program control.

It is a block diagram which shows the structure of the video signal processing apparatus of Embodiment 1 of this invention. It is a block diagram which shows the structure of an example of the field state determination means 14 of FIG. It is a block diagram which shows the structure of an example of the telecine determination means 20 of FIG. It is a block diagram which shows the structure of an example of the telecine sequence phase detection means 212 of FIG. It is a block diagram which shows the structure of an example of the telecine detection signal generation means 215 of FIG. It is a block diagram which shows the structure of an example of the sequence deviation detection means 22 of FIG. It is a figure which shows the relationship with the video signal before conversion of the telecine video signal in the telecine conversion by 2-3 pulldown system. It is a figure which shows the signal which appears in each part of the telecine detection means 6 of FIG. 1 when the telecine video signal shown in FIG. 7 is input. FIG. 8 is a diagram showing a relationship with a video signal before conversion when the regularity of the field sequence is disturbed due to editing for broadcasting, insertion of CM or telop, etc. in the middle of the sequence of the telecine video signal shown in FIG. is there. It is a figure which shows the signal which appears in each part of the telecine detection means 6 of FIG. 1 when the telecine sequence video signal shown in FIG. 9 is input. It is a block diagram which shows the structure of the sequence shift | offset | difference detection means 22 used in Embodiment 2 of this invention. It is a figure which shows the signal which appears in each part of the telecine detection means 6 of Embodiment 2 when the telecine video signal shown in FIG. 9 is input. In the case where edit points exist in the sixth to ninth fields Cb, Ct, Cb, and Ct in the middle of the telecine video signal, four fields are inserted as still images, and the regularity of the sequence is disturbed. It is a figure which shows the signal which appears in each part of the telecine detection means.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Timing signal generation means 2, 3 Field delay means 4 Motion detection means 5 Motion adaptive interpolation calculation means 6 Telecine detection means 7 Telecine interpolation processing means 8 Selection means 9 Synthesis means 11 Pixel difference detection means 12 Difference comparison means 13 Predetermined area accumulation counter 14 Field state determination means 20 Telecine determination means 21 Telecine sequence detection means 22 Sequence deviation detection means 30 Correlation determination means 40 Scan line interpolation means 45 Motion adaptive interpolation Processing means 211 moving picture field counter 212 telecine sequence phase detecting means 213 moving picture field count value comparing means 214 still image field continuous detecting means 215 telecine detection signal generating means 221 still image field counter 222 count value comparing hand Stage, 223 delay means, 224 logic gate, 225 deviation determination means, 100 input terminal, 101, 102 terminal, 103 output terminal.

Claims (16)

A video signal processing device that generates an interpolated scanning line from an input interlaced video signal and converts it to a progressive video signal,
Field delay means for delaying a video signal by a plurality of fields in a field unit;
A correlation between the video signal of the field of interest and the video signal of the field one frame before the field of interest is detected, and the field of interest is a still image field or a moving image field based on the strength of the correlation between frames. Correlation determination means for determining whether or not
Based on the output signal from the correlation determining means, a field sequence of telecine conversion in a video signal telecine-converted by the 2-3 pull-down method is detected, and the input video is detected from the detection result of the field sequence of the telecine conversion. Telecine sequence detection means for determining whether the signal is a telecine converted video signal;
Based on the output signal from the correlation determining means and the detection result of the telecine sequence detecting means, the field is different from the field of the specific phase to be the still image field in the telecine conversion field sequence detected by the telecine sequence detecting means. Sequence deviation detecting means for detecting this when the is a still image field,
Scanning line interpolation means for switching processing for generating an interpolation scanning line signal based on the detection result from the telecine sequence detection means,
The video signal processing apparatus, wherein the telecine sequence detection means determines whether the video signal is the telecine-converted video signal using the output of the sequence deviation detection means.   The correlation determination means is determined based on a difference value or a difference value between each pixel in a predetermined area or all areas in the target field of the video signal and a pixel at the same position in the field one frame before. 2. The video signal processing apparatus according to claim 1, wherein whether the field of interest is a still image field or a moving image field is determined based on whether or not the number of pixels whose value is greater than a predetermined value is greater than or equal to a predetermined number. The correlation determination means includes
Pixel difference for detecting a value determined based on a difference value or a difference value from a pixel at the same position in a field before one frame for each pixel in a predetermined area or all areas in the target field of the video signal Detection means;
A difference comparison unit that compares the output of the pixel difference detection unit with a predetermined value and determines the presence or absence of movement in each pixel;
Based on the output from the difference comparison means, the cumulative number of pixels determined to have movement or the number of pixels determined to have no movement is counted for pixels in a predetermined area or all areas in the field. A counter,
Field state determination means for comparing the count value of the cumulative counter with a predetermined value and determining whether the field of interest is a still image field or a moving image field from the comparison result;
3. The video signal processing apparatus according to claim 2, wherein an output signal of the field state determination unit is used as an output of the correlation determination unit. The field state determination means includes
First comparison means for comparing the count value of the cumulative counter with a first predetermined value and outputting a first comparison result signal indicating the comparison result;
A second comparing means for comparing the count value of the cumulative counter with a second predetermined value and outputting a second comparison result signal indicating the comparison result;
4. The video signal processing apparatus according to claim 3, wherein the output signals of the first and second comparison means are output from the field state determination means. 5. The telecine sequence detecting means is
When the sequence deviation detecting means detects that a field different from the field of the specific phase is a still image field, a signal indicating that the video signal has been telecine converted is not output. The video signal processing apparatus according to claim 1. The telecine sequence detection means determines whether or not the input video signal is a telecine-converted video signal based on an output from the correlation determination means for five fields of the input video signal. And
A moving picture field counter for detecting a continuous number of fields in which the output signal from the correlation determining means indicates a moving picture field;
Still image field continuous detection means for detecting a continuous number of fields in which an output signal from the correlation determination means indicates a still image field;
Telecine sequence phase detection means for detecting the phase of the telecine conversion field sequence in the video signal telecine-converted by the 2-3 pull-down method based on the output signal from the correlation determination means and the output from the video field counter;
Based on the output signal from the correlation determination means and the output from the video field counter, find the number of repetitions of the detected field sequence of the telecine conversion,
Based on the number of repetitions, a telecine detection signal generating means for detecting whether or not the sequentially input video signal has been subjected to telecine conversion,
The output of the telecine detection signal generation means is output as a signal indicating whether or not it is a telecine converted video signal, and the output of the telecine sequence phase detection means is output as a signal indicating the phase of the telecine conversion field sequence. The video signal processing apparatus according to claim 5. The telecine sequence phase detecting means is
Specific phase field detection means for detecting the timing of a field that should become a still image field after four consecutive video fields based on an output signal from the correlation determination means and an output from the video field counter;
Phase signal generating means for generating a repetitive signal that repeats in a five-field period in synchronization with the field detected by the specific phase field detecting means,
The video signal processing apparatus according to claim 6, wherein the output of the phase signal generating means is output as a signal indicating a phase of a field sequence which is a 2-3 pulldown conversion phase. Furthermore, a moving image field count value comparing means for comparing the output from the moving image field counter with a predetermined value,
The telecine detection signal generating means is
A reset condition determining means for determining a reset condition based on an output of the moving image field count value comparing means, an output of the still image field continuous detecting means, and an output of the sequence deviation detecting means;
Based on the output signal of the correlation determination unit and the output of the moving image field counter, the number of consecutive field sequences that become a still image field after four consecutive moving image fields is obtained based on the output signal of the correlation determining unit and the output of the moving image field counter. Field sequence continuous number detection means;
Field sequence continuity number comparing means for comparing the continuity number obtained by the field sequence continuity detecting means with a predetermined value and outputting a signal indicating the comparison result; and
7. The video signal processing apparatus according to claim 6, wherein an output signal of the field sequence continuous number comparison means is output as a signal indicating whether or not the video signal has been telecine converted. The field sequence continuous number detecting means includes:
When the output signal of the correlation determination means indicates a still image field and the output of the moving image field counter indicates a predetermined value,
The video signal processing apparatus according to claim 8, wherein the count value is increased by one. The reset condition determining means includes
When the output of the moving image field count value comparison means indicates that the number of continuous moving image fields is equal to or greater than a predetermined value, the output of the still image field continuous detecting means outputs a predetermined number of continuous still image fields. When it is indicated that the above condition is satisfied, the output of the sequence deviation detecting means satisfies the condition when any of the conditions when the sequence deviation is indicated is satisfied. 9. The video signal processing apparatus according to claim 8, wherein the means is reset. The sequence deviation detecting means is
It operates according to the output of the telecine sequence detection means, and counts the number of still image fields detected by the correlation determination means in fields other than the specific phase field in the 5-field field sequence in 2-3 pulldown. Image field counting means;
Comparing means for comparing the count value of the still image field counting means with a predetermined value,
The video signal processing apparatus according to claim 1, wherein a signal indicating a comparison result of the comparison unit is output as a signal indicating a detection result of the sequence deviation detection unit. The sequence deviation detecting means is
It operates according to the output of the telecine sequence detection means, and counts the number of still image fields detected by the correlation determination means in fields other than the specific phase field in the 5-field field sequence in 2-3 pulldown. Image field counting means;
Generates a signal indicating whether the output signal of the correlation determination unit is a moving image field and whether the count value of the still image field counting unit one field before is greater than or equal to a predetermined value And a deviation presence / absence judging means for
The video signal processing apparatus according to claim 1, wherein a signal indicating a determination result of the shift presence / absence determination unit is output as a signal indicating a detection result of the sequence shift detection unit. The scanning line interpolation means includes
Intra-field interpolation that generates an interpolation signal by calculation processing based on a video signal that represents a value of a neighboring pixel in the interpolation target field for the video signal in the interpolation target field, or an interpolation target field for the video signal in the interpolation target field Motion adaptive interpolation processing means for performing inter-field interpolation to generate an interpolation signal by arithmetic processing based on video signals of adjacent fields of
Telecine interpolation processing means for generating an interpolation signal based on a video signal one field before or one field after the interpolation target field in accordance with a signal indicating a field sequence of telecine conversion from the telecine sequence detection means;
A selection means for selecting and outputting either the interpolation signal by the motion adaptive interpolation processing means or the interpolation signal by the telecine interpolation processing means according to the detection result in the telecine sequence detection means;
When it is detected by the detection result by the telecine sequence detection means that the video signal is telecine converted, the interpolation signal by the telecine interpolation processing means is selected, and otherwise, the motion adaptive interpolation processing means Select the interpolation signal,
2. The video signal processing apparatus according to claim 1, wherein the selected interpolated scanning line signal is output. A video signal processing method for generating an interpolated scanning line from an input interlaced video signal and converting it to a progressive video signal,
A correlation between the video signal of the field of interest and the video signal of the field one frame before the field of interest is detected, and the field of interest is a still image field or a moving image field based on the strength of the correlation between frames. A correlation determination step for determining whether or not
A sequence detection step of detecting a field sequence of telecine conversion in the video signal telecine-converted by the 2-3 pull-down method based on the determination result in the correlation determination step;
A telecine determination step for determining whether the input video signal is a telecine-converted video signal from the detection result of the telecine conversion field sequence;
Based on the result of the determination in the correlation determination step and the result of the determination in the telecine sequence determination step, the field of the specific phase to be a still image field in the telecine conversion field sequence detected in the sequence detection step is different. A displacement detection step for detecting this when the field is a still image field;
A scanning line interpolation step for switching processing for generating an interpolation scanning line signal based on the detection result in the sequence detection step and the determination result in the telecine determination step,
The video signal processing method, wherein the determination in the telecine determination step is performed also using the detection result of the shift detection step. In the deviation detection step,
The number of still image fields detected in the correlation determination step is counted in fields other than the specific phase field in the field sequence of 5 fields in 2-3 pulldown detected in the phase detection step. The video signal processing method according to claim 14, wherein when the value is equal to or greater than a predetermined value, it is determined that the sequence has shifted. In the deviation detection step,
The still image field detected in the correlation determination step is counted in a field other than the specific phase field in the field sequence of 5 fields in 2-3 pulldown detected in the phase detection step.
When it is determined in the correlation determination step that the field is a moving image field, and the count value of the still image field one field before is greater than or equal to a predetermined value,
The video signal processing method according to claim 14, wherein it is determined that the sequence has shifted.

Images